Interpretable machine learning framework reveals microbiome features of oral disease.
Yueyang Yan1, Xin Bao2, Bohua Chen3
1Key Laboratory for Zoonoses Research of the Ministry of Education, Institute of Zoonosis, College of Veterinary Medicine, Jilin University, Changchun 130062, China.
Microbiological Research
|September 20, 2022
Summary
This study identified key oral microbiome features linked to oral diseases using machine learning. These findings support lifestyle interventions for oral disease prevention and personalized medicine approaches.
Area of Science:
- Microbiology
- Genomics
- Computational Biology
Background:
- The oral microbiome is crucial in oral disease development, but specific disease-associated microbes are poorly understood.
- Characterizing the oral microbiome is essential for understanding and preventing oral diseases.
Purpose of the Study:
- To identify specific oral microbiome features associated with oral diseases.
- To develop a machine learning framework for analyzing complex microbiome data.
- To explore the potential for lifestyle interventions in oral disease prevention.
Main Methods:
- Collected saliva samples from 140 individuals.
- Performed 16S amplicon sequencing for microbiome analysis.
- Developed an interpretable machine learning framework and utilized SHapley Additive exPlanations (SHAP) to construct Microbiome Risk Scores (MRSs).
Main Results:
- Identified 14 distinct oral microbiome features associated with oral diseases.
- Developed Microbiome Risk Scores (MRSs) using SHAP values.
- Found correlations between MRSs and individual physiological indicators and lifestyle habits.
Conclusions:
- The study successfully identified oral microbiome features linked to oral diseases.
- Demonstrated the potential for lifestyle interventions to prevent oral diseases.
- Provided a reference method for precision medicine in oral health.
